The invention concerns a method for determining the mass of a vehicle driven by at least one prime mover.
EP 0 666 435 A2 has disclosed such a method which comprises two measurements, offset in time, of the traction force produced by the prime mover and the acceleration resulting therefrom. It is assumed that the substantially the same in both measurements so that the unknown variables can be abridged. In this known method, while one clutch is disengaged for the purpose of a gear shift of a stepped variable speed transmission, both during a traction phase and during a traction-free phase a value of a gear torque and a value of a vehicle acceleration are each determined and the actual mass of the vehicle can be calculated therefrom. Such a method can advantageously be used within a driving strategy in vehicles having automated transmissions in order to be able to calculate from a gear shift, for ex., whether the traction in the new speed is still sufficient, the load state and thus the vehicle mass having a decisive influence. A great precision of the method is necessary for such applications.
It is advantageous in this known method that it cannot automatically start in the rear unnoticeably for the driver. During a change of gear of a stepped variable speed transmission, the traction-free phase and traction phase states, which are produced in any case, are used for the measurements defined as marginal conditions. The disadvantage of this method is that the values of the gear torques or of the accelerations used for the calculation are subject to scatterings determined by the measuring technique and, therefore, leading to inaccurate results.
The method described in WO 93/18375 likewise refers to the principle of two measurements of traction and movement variables offset in time. In order to eliminate any gravitational influences, it is proposed that the driver twice traverses an identical distance with a different traction input to carry out the measurement. This can be carried out only with difficulty due to practical problems when using the vehicle in real street traffic. It is alternatively proposed that a distance, with a precisely known profile, be run through. This is also problematic, since such a distance is not available to every driver and the expenditure for such a measurement is not acceptable for practical utilization.
The problem on which this invention is based is, therefore, to develop a method for determining a mass of a vehicle that, without effort for the driver, starts automatically, is not detectably by the driver and at the same time achieves more precision than a known method.
The method is, in particular, suited to automated transmissions in which the traction is interrupted during a change of gear.
During the two data collection periods, the duration can be different, but always longer than a specific minimum duration and of one is within the traction-free phase during a gearshift and the other is during a traction phase before or after the gearshift. The values of the traction variable and of the movement variable are determined with great accuracy with the result of an altogether high precision method.
The method is based on the following equation:       M    Fzg    =                              ∫          t0          t1                ⁢                              F            Zug                    ⁢                      xe2x80x83                    ⁢                      ⅆ            t                              -                        M          Gang                ⁢                  (                                    v              1                        -                          v              0                                )                            (                        v          1                -                  v          0                -                  v          3                +                  v          2                    )      
Wherein:
The method is based on the following equation:
The integral about the traction from t2 to t3 does not appear in this equation, since Fzug is zero during the force-free phase.
It has been demonstrated that the results can be substantially improved by the adjustment factor MGang, since the drag torques on the input side, especially for low gear steps, have a considerable influence. A form of this equation, which is adequate for programmation of an electronic transmission control and in which the integral contained in the equation is replaced by an approximation method of discrete values in time, is given by the following equation:             ms_fzg      ⁢      _akt        =                            I          01                -                  ms_fzg          ⁢                      _korr            ·                          (                                                v_fzg                  ⁢                  _filt                  ⁢                                      (                                          k                      1                                        )                                                  -                                  v_fzg                  ⁢                  _filt                  ⁢                                      (                                          k                      0                                        )                                                              )                                                                                      (                                                v_fzg                  ⁢                  _filt                  ⁢                                      (                                          k                      1                                        )                                                  -                                  v_fzg                  ⁢                  _filt                  ⁢                                      (                                          k                      0                                        )                                                  -                                                                                        k_t0123              ·                              (                                                      v_fzg                    ⁢                    _filt                    ⁢                                          (                                              k                        3                                            )                                                        -                                      v_fzg                    ⁢                    _filt                    ⁢                                          (                                              k                        2                                            )                                                                                                                          I      01        =          #      ⁢              t_abt        ·        #            ⁢      k_korr      ⁢              _ms        ·                                            ∑              k                                      0              +              1                                      k              1                                ⁢                      xe2x80x83                    ⁢                      (                          f_zs              ⁢                              (                k                )                                      )                              
Wherein:
It is apparent that the integral about the traction, during the traction phase (see first equation), can be approximated with more precision by using the method of discrete values in time. The speeds at the beginning and the end of the data collection periods likewise can be well controlled so that a very high precision can be achieved.
The method is especially adequate for power shiftable transmissions. In the transmissions, the number of gears is usually lower from which it follows that the ratio ranges are larger for consecutive gears.
Therefore, immediately before a gear shift, the traction in many cases is clearly different from the traction immediately after the gear shift. The method uses these traction states abruptly following each other with the first data collection period being immediately before a gear shift and the second data collection period being immediately after the gear shift. The beginning of the first data collection period can be triggered, e.g. by a shift command. The time between a shift command and a start of the gear shift, usually needed to fill the clutches to be engaged, can thus be used for the first measurement.
The following equation serves to calculate the vehicle mass:       M    Fzg    =                                                                        ∫                                  t                  0                                                  t                  1                                            ⁢                                                F                  Zug                                ⁢                                  xe2x80x83                                ⁢                                  ⅆ                  t                                                      -                                          ∫                                  t                  2                                                  t                  3                                            ⁢                                                F                  Zug                                ⁢                                  xe2x80x83                                ⁢                                  ⅆ                  t                                                      -                                                                          M              Gang                        ,                                          t01                ⁢                                  (                                                            v                      1                                        -                                          v                      0                                                        )                                            +                              M                Gang                                      ,                          t23              ⁡                              (                                                      v                    3                                    -                                      v                    2                                                  )                                                                (                        v          1                -                  v          0                -                  v          3                +                  v          2                    )      
Wherein:
The integral about the traction of t2 to t3 is taken into account in this equation since a traction is close during both data collection periods. The conversion to an approximation method of discrete values in time accordingly can be as in the method of claim 1.